A data warehouse is used to acquire information from a database, and a configured task in the data warehouse is corresponding to one or more data processing steps. Characteristics such as unidirectional dependence, independence, or a time constraint may exist between tasks. A task scheduling system is an important component of the data warehouse, and is used to schedule computing resources in the data warehouse to implement processing of the configured tasks.
An existing task scheduling process in a data warehouse is as follows: A task scheduling system drives execution of a task according to a set configured in a task deployment system by data development personnel, dependence between tasks in the set, and the like, and then after execution of each task is completed, a next task is executed sequentially or according to the dependence between the tasks, without considering a quantity of resources required by the next task, which results in low resource utilization and poor processing performance in a task executing process.